> For the complete documentation index, see [llms.txt](https://thecontentforge.gitbook.io/thecontentforge-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://thecontentforge.gitbook.io/thecontentforge-docs/feature-glossary/glossary/image-forge.md). # Image Forge *Last updated: 2026-08-07.* Image Forge is where visual assets get made: a full design editor plus a set of one-job utilities for generating, resizing, upscaling and cleaning up images. The area header carries a Beta badge in the product. Short links and QR codes live in this same area and have their own page, Links and QR. * **Image Editor** - A layered design canvas with text, shapes, lines, images, drawing and erasing tools, an eyedropper, crop, rulers, edge and centre snapping, an optional grid, zoom, undo and redo, a searchable command palette and keyboard shortcuts. Photoshop files and vector artwork import as editable layers rather than flat pictures. It is the workspace where a social graphic, thumbnail or banner actually gets built, so a team can produce finished assets without opening a separate design tool. * **Layers** - A layer list down the side of the editor for showing and hiding, locking, reordering and deleting each element, with per-layer opacity and blend mode, plus align, duplicate, flip and rotate on the selected layer. Nothing is baked in until you export, so a headline or a logo can be moved or removed at any point. * **Blank Canvas** - Starts a new project at a chosen size, either from a named preset for the common social and video formats or from custom dimensions, with a transparent, solid or custom background. Custom sizes can be saved as a reusable preset kept private or shared with the whole team, so recurring formats never have to be re-typed. * **Thumbnail Studio** - Generates a video thumbnail from a title and a few choices: format, visual style, caption handling, and whether to include an on-camera host with a chosen expression and side of frame. The finished image opens straight into the editor as the base layer of a new project, ready for text, badges and other layers on top. * **Host Library** - Saves the faces of your on-camera people as reusable assets shared across the team, or lets you attach a one-off image for a single thumbnail. The host you used last time is remembered, so everyone generates with the same faces instead of re-uploading them each time. * **Resize Image** - Takes an existing image and refits it to any output size with real framing control: contain, cover and crop, stretch, pad or a smart crop suggestion, plus zoom, pan, rotate, flip, background fill, output format and quality. A live preview with thirds and centre guides shows the result before you run it, and undo and redo apply to the framing itself. * **Upscale Image** - Increases resolution and detail on an existing image, with a 2x or 4x choice and a mode tuned for text and interface screenshots, logos and icons, or thumbnails and social art. The tool shows the finished pixel size before you run it and greys out a scale that would be too large, and it tells you plainly when upscaling is temporarily unavailable rather than failing silently. * **AI Background** - Generates a background image from a written description and drops it into the current design as a layer, with the output also saved to your asset library. Useful when you need something abstract or on-theme and stock imagery would look generic. * **AI Canvas** - Select any region of the canvas as a rectangle, ellipse or freehand lasso and then regenerate it: fill it with something new from a prompt, erase an object and blend the surroundings back in, repair a damaged area, or invert the selection to keep one thing and regenerate everything around it. There is a before and after comparison view and the last selection can be reused, so refining a result does not mean redrawing the mask. * **Background Removal and Edge Refinement** - Removes a background with a mode hint for logos, products, people, photos or screenshots, plus a higher-quality pass, a tolerance-based magic wand, defringe and matte removal, and edge contract, expand, feather, despeckle and hole-fill controls. The cutout can be previewed against a checkerboard, white, black, red or blue backdrop so a halo shows up before you place it. * **Paint Bucket and Colour Replace** - Fills an area with a chosen colour at an adjustable tolerance and opacity, either contiguous or across the whole layer, and swaps one colour for another across an image while optionally keeping the original lighting. Either can write to a new layer instead of altering the original. * **Image Utilities** - Per-layer one-click operations on an image: trim to content, split into a two by two or three by three tile set, key out a colour, fix broken transparency, apply a gradient map, diff two selected images, stitch or collage them, and produce an app icon pack. These are the small fixes that otherwise send someone back to a desktop editor. * **Icons and Stickers** - A searchable built-in library of vector glyphs covering arrows and calls to action, social and engagement marks, news and warning symbols, charts and finance, crypto and badges. They insert as real vectors, so they recolour with the normal fill control and stay sharp at any size. * **Templates** - A searchable, categorised library of ready-made layouts for social posts, news and breaking-news cards, quotes and testimonials, stats and charts, crypto and market updates, launches, events, app screenshots and carousels. Any design you make can be saved as your own template and starred as a favourite, so a house style becomes repeatable. * **Colour System and Brand Colours** - Applies colour by role across fills, strokes, text and backgrounds, with suggested, recent, project, saved and brand palettes in one place, plus gradients and a palette generator that pulls colours out of the selected image. Designs stay on brand without anyone hunting for hex codes. * **Adjustments, Effects and Filters** - Per-layer brightness, contrast, saturation, vibrance, hue, gamma, blur and grain with a live histogram, one-click looks such as vivid, noir, warm, cool and vintage, and text and object effects including glows, outlines, drop shadows and sticker strokes. * **Marketing Elements** - Drops in brand-aware call-to-action badges, lower thirds, watermarks, "NEW" and "LIVE" badges and price tags as fully editable layers, so promotional furniture is one click rather than a rebuild every time. * **Recipes** - One-click chains that run several steps together, such as auto-enhance, a clean cutout, cutout straight to a transparent file, add watermark, export and save, an export pack, compress, and thumbnail variants. Frequently repeated sequences also appear as saved workflows with a run count and the outcome of the last run, which cuts the busywork out of routine production. * **Export** - Saves the finished design as PNG, transparent PNG, WebP, transparent WebP, JPG or vector SVG, or straight into the asset library. * **Send To** - Pushes a finished design directly to Content Forge, or saves it as a transparent overlay that Video Forge can pick up, so an image moves into the post or the video without a download-and-re-upload round trip. * **Save to Asset Library** - Every export, generated image and utility output can be saved to the shared asset library, recorded with the tool that produced it and the source image it came from. That lineage means the whole team can reuse the asset later and see where it originated. * **Projects and Autosave** - Work is saved as a named project that autosaves as you edit and shows a picture of the current design in the project grid, so finding the right one does not depend on remembering its name. Projects can be reopened, duplicated through the gallery or deleted. * **AI Gallery** - A running strip of your recent generations with the prompt, mode and size, where you can star one for yourself, remix it into a new project, download it, or pick two and open them together in one project to compare. * **Batch Image Transforms** - Run from the shared asset library: select any number of images, choose a pack, and queue them all at once. The packs are a social resize pack that produces square, portrait, story and landscape versions of each image, a thumbnail variants pack at small, medium and large sizes, a logo cleanup pass that trims dead space and keeps the result at a sensible size, and a compressed web-optimised pack. Finished images land back in the library when they are ready, each recording which original it came from. * **Image Job Queue** - A quiet indicator shows when image work is running anywhere in your workspace, so nobody re-submits a job that is already in progress. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://thecontentforge.gitbook.io/thecontentforge-docs/feature-glossary/glossary/image-forge.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.